Programming

deploy-rs kept deploying our infrastructure while its maintenance backlog grew. This year, we returned to that backlog and tagged version 1.0.0, the first versioned release of the Nix deployment tool we introduced in 2020. Here’s what changed, how we use it to deploy systems and individual services, and what we plan to tackle next.

GHC now has a policy for AI-assisted contributions. Contributors can use LLMs, but they remain responsible for everything they submit, must disclose directly generated material, and are expected to keep the review burden on maintainers under control. In this article, we look at what the policy says, how it compares with the approaches taken by LLVM, OpenJDK, GCC, Zig, and other projects, and why we think GHC has landed on a reasonable set of rules for AI-assisted work.

Rust’s infrastructure runs on limited engineering time. Saving compute can mean adding maintenance work, and reviewing contributions competes with keeping services running. In this interview, Marco Ieni of the Rust Foundation explains how the team reduced CI costs and why it sometimes accepts duplicated work for a simpler system. We also discuss running untrusted code, infrastructure as code, release-plz, and what it takes to support Rust’s growing ecosystem.

Not so long ago, LLMs were incapable of producing output that matched the quality of a competent human. Hence the derogatory term “slop”. Like it or not, this is no longer always the case. You can g…

Most blockchains are built around transparency: smart contracts, transactions, and state changes are visible to everyone. That works well in many cases, but it becomes a problem when organizations need to coordinate without exposing sensitive data. The Canton Network takes a different approach. It is designed for privacy, data sovereignty, and interoperability, allowing participants to share only the information required for a given transaction. Canton applications are commonly built with Daml, a strongly typed functional smart contract language centered on parties, authorization, contracts, and multi-party workflows. In this article, we’ll introduce Canton and Daml, compare Daml with Solidity, and build a small banking application. Along the way, we’ll see how privacy and authorization requirements influence smart contract architecture.

Over the past years, Serokell has worked on several parts of the Haskell ecosystem: GHC development, language proposals, Dependent Haskell, developer tooling, and education. In this article, we’ll go through the main contributions and explain why they matter.